gpt-oss-120b vs Phi-4 API pricing
Updated
gpt-oss-120b: USD 0.03 input, USD 0.17 output per 1M tokens (CoreWeave Forge). Phi-4: USD 0.07 input, USD 0.14 output per 1M tokens (DeepInfra and 1 other). At 3 input tokens per output token, Phi-4 costs 35% more than gpt-oss-120b (USD 0.0875 vs USD 0.065 per 1M).
How do gpt-oss-120b and Phi-4 differ on price?
- Cheapest input price: gpt-oss-120b USD 0.03 against USD 0.07 per 1M tokens, so Phi-4 costs 2.3× as much.
- Cheapest output price: Phi-4 USD 0.14 against USD 0.17 per 1M tokens, so gpt-oss-120b costs 1.21× as much (21% more).
- Among the 201 indexed models, gpt-oss-120b ranks 9th cheapest and Phi-4 14th on the 3:1 blend.
- The cheaper model changes at 0.75 input tokens per output token.
- Only gpt-oss-120b lists a cached-input discount: gpt-oss-120b at 4 standard prices of 21, from USD 0.015 (Fireworks AI, 90% below that provider's input price). Phi-4's cached tokens bill at the input price.
- Both state a context window of up to 131,072 tokens.
- gpt-oss-120b has 21 providers with a standard price, 18 more than Phi-4; 3 providers sell both (DeepInfra, Featherless and Microsoft Azure).
- On one account, DeepInfra charges USD 0.07025 for gpt-oss-120b (1.08× its cheapest price) and USD 0.0875 for Phi-4 (its cheapest price) at 3:1, the lowest combined price among the providers that sell both.
- Shopping around matters more for gpt-oss-120b: its dearest standard price is 6.9× its cheapest, against 2.5× for Phi-4.
- gpt-oss-120b has a batch price at 2 providers, from USD 0.07875 at 3:1 (Google Vertex AI); Phi-4 has no batch price on file.
- For the one-developer coding agent preset, the gap is USD 2,408 a year in gpt-oss-120b's favour (gpt-oss-120b USD 69.30 a month, Phi-4 USD 270.00).
- Per character of input, the gap depends on the language: gpt-oss-120b costs 57% less than Phi-4 on Code text but 86% less than Phi-4 on Hindi, because the two tokenizers split those languages differently.
gpt-oss-120b and Phi-4 prices side by side
USD per 1M tokens, standard tier. "Cheapest" is the lowest price at 3 input tokens to 1 output token; each line can come from a different provider.
| Price | gpt-oss-120b | Phi-4 |
|---|---|---|
| Model page | OpenAI GPT OSS 120B OpenAI · 21 providers | Phi-4 Microsoft · 3 providers |
| Cheapest standard price, blended 3:1 | USD 0.065 lower USD 0.03 in / USD 0.17 out at CoreWeave Forge CoreWeave Forge page checked 2026-11-15 | USD 0.0875 USD 0.07 in / USD 0.14 out at DeepInfra and 1 other DeepInfra page checked 2026-11-14 |
| Cheapest input | USD 0.03 lower CoreWeave Forge source | USD 0.07 DeepInfra source |
| Cheapest cached input | USD 0.015 Fireworks AI source | No standard price lists a cache price; cached tokens bill at the input price. |
| Cheapest output | USD 0.17 CoreWeave Forge source | USD 0.14 lower DeepInfra source |
| Maker's own API | OpenAI publishes no paid standard price on its own API that we found. | Microsoft publishes no paid standard price on its own API that we found. |
| Median standard price | USD 0.10 / USD 0.60 input / output, 21 prices | USD 0.07 / USD 0.14 input / output, 3 prices |
| Context window | 131,072 tokens, largest stated; from 128,000 at some providers | 131,072 tokens, largest stated; from 16,384 at some providers |
| Providers with a standard price | 21 | 3 |
- Lower marks the lower price on a line; tie, equal prices. Other tiers and free rows are on the model pages.
- Promo: the provider calls the price promotional or discounted, so it is current now and can end without notice.
Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).
What do gpt-oss-120b and Phi-4 cost for a real workload?
30 days at each model's cheapest standard price for the workload (presets of the cost calculator; cached input at the cached price where listed).
| Workload | gpt-oss-120b | Phi-4 | Difference |
|---|---|---|---|
| Support chatbot 2,000 requests a day, 1,500 input tokens (60% cached), 300 output | USD 5.76 CoreWeave Forge | USD 8.82 DeepInfra | USD 3.06 gpt-oss-120b lower |
| RAG over documents 1,000 requests a day, 8,000 input tokens (30% cached), 500 output | USD 9.75 CoreWeave Forge | USD 18.90 DeepInfra | USD 9.15 gpt-oss-120b lower |
| Coding agent (one developer) 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning | USD 69.30 CoreWeave Forge | USD 270.00 Microsoft Azure | USD 200.70 gpt-oss-120b lower |
gpt-oss-120b is cheaper on all three workloads. The widest gap is on Coding agent (one developer): the dearer model costs 3.9× as much as the cheaper one.
Does the cheaper model change with your input:output ratio?
The cheaper model changes with the mix: below 0.75 input tokens per output token, Phi-4 is cheaper; above 0.75, gpt-oss-120b is cheaper. At the site's 3:1 blend, gpt-oss-120b is cheaper. More input per output token, as with retrieval or agents that resend long contexts, favours the lower input price; long answers favour the lower output price.
| Input : output tokens | gpt-oss-120b | Phi-4 |
|---|---|---|
| 1 : 3 (long answers) | USD 0.135 CoreWeave Forge | USD 0.1225 lower DeepInfra |
| 1 : 1 | USD 0.10 lower CoreWeave Forge | USD 0.105 DeepInfra |
| 3 : 1 (the site’s blend) | USD 0.065 lower CoreWeave Forge | USD 0.0875 DeepInfra |
| 10 : 1 (retrieval, long prompts) | USD 0.04273 lower CoreWeave Forge | USD 0.07636 DeepInfra |
| 30 : 1 (agents, big contexts) | USD 0.03452 lower CoreWeave Forge | USD 0.07226 DeepInfra |
USD per 1M tokens at the cheapest standard provider for each mix, before cache discounts. Thinking tokens bill as output.
What do gpt-oss-120b and Phi-4 cost per character in other languages?
Cost to send 1M characters at each model's cheapest non-promotional standard input price, with the gpt-oss (o200k_harmony) tokenizer for gpt-oss-120b and the Phi-4 tokenizer for Phi-4:
| Content | gpt-oss-120b | Phi-4 |
|---|---|---|
| English | USD 0.0064 lower CoreWeave Forge | USD 0.015 DeepInfra |
| Arabic MSA | USD 0.0096 lower CoreWeave Forge | USD 0.050 DeepInfra |
| Egyptian Arabic | USD 0.0100 lower CoreWeave Forge | USD 0.049 DeepInfra |
| Hindi | USD 0.0097 lower CoreWeave Forge | USD 0.069 DeepInfra |
| Chinese | USD 0.023 lower CoreWeave Forge | USD 0.082 DeepInfra |
| Spanish | USD 0.0068 lower CoreWeave Forge | USD 0.019 DeepInfra |
| Code | USD 0.0076 lower CoreWeave Forge | USD 0.018 DeepInfra |
From the real cost by language dataset (measured tokens per character on FLORES-200 and other texts). Your own text will differ.
Check it on your own traffic
- OpenAI GPT OSS 120B API pricing: all 21 providers and other tiers.
- Phi-4 API pricing: all 3 providers and other tiers.
- Cost calculator with gpt-oss-120b and Phi-4 preloaded
- gpt-oss token counter
- Phi-4 token counter
- Prompt caching calculator
More comparisons: gpt-oss-120b vs gpt-oss-20b · gpt-oss-120b vs Llama 3.3 70B · Nemotron 3 Super vs gpt-oss-120b · gpt-oss-120b vs DeepSeek V4.1 Flash · GPT-6 Luna vs gpt-oss-120b · gpt-oss-120b vs Qwen3.8 27B · gpt-oss-120b vs Gemma 4 26B A4B · every gpt-oss-120b comparison · every Phi-4 comparison · all 662 price comparisons
Questions
Is gpt-oss-120b cheaper than Phi-4?
Yes. On a blend of 3 input tokens to 1 output token, gpt-oss-120b costs USD 0.065 per 1M tokens and Phi-4 USD 0.0875, at each model's cheapest standard price. For input alone, gpt-oss-120b is cheaper (USD 0.03 against USD 0.07 per 1M); for output alone, Phi-4 is (USD 0.14 against USD 0.17). This compares price only, not quality.
What do gpt-oss-120b and Phi-4 cost per 1M tokens?
gpt-oss-120b: from USD 0.03 input and USD 0.17 output at CoreWeave Forge; median of 21 standard prices USD 0.10 / USD 0.60. Phi-4: from USD 0.07 input and USD 0.14 output at DeepInfra and 1 other; median of 3 standard prices USD 0.07 / USD 0.14. Each price is read from the provider's own pricing page; the source is linked in the table.
Which providers sell both gpt-oss-120b and Phi-4?
3 providers have a standard price for both: DeepInfra (USD 0.07025 and USD 0.0875 at 3:1), Featherless (USD 0.2625 and USD 0.0875 at 3:1), Microsoft Azure (USD 0.2625 and USD 0.2188 at 3:1). The first is the lowest combined price. Buying each model from its own cheapest provider costs USD 0.065 and USD 0.0875.
Which is cheaper for a coding agent or a chatbot, gpt-oss-120b or Phi-4?
For a coding agent (1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning): USD 69.30 a month on gpt-oss-120b (CoreWeave Forge) against USD 270.00 on Phi-4 (Microsoft Azure), a difference of USD 200.70. For a support chatbot (2,000 requests a day, 1,500 input tokens (60% cached), 300 output): USD 5.76 a month on gpt-oss-120b (CoreWeave Forge) against USD 8.82 on Phi-4 (DeepInfra), a difference of USD 3.06. Each uses the cheapest standard price for that workload, cached input included. Try your own numbers.
At what input:output ratio does the cheaper of gpt-oss-120b and Phi-4 change?
The cheaper model changes with the mix: below 0.75 input tokens per output token, Phi-4 is cheaper; above 0.75, gpt-oss-120b is cheaper. At the site's 3:1 blend, gpt-oss-120b is cheaper. More input per output token, as with retrieval or agents that resend long contexts, favours the lower input price; long answers favour the lower output price.
Do gpt-oss-120b and Phi-4 have cached input pricing?
gpt-oss-120b from USD 0.015 per 1M cached input tokens (Fireworks AI; 4 of 21 standard prices list a discount). Phi-4: no standard price lists a cached-input discount. Cached tokens are billed at the input price where no cache price is published.
Which has the larger context window, gpt-oss-120b or Phi-4?
Both go up to 131,072 tokens on the providers that state it.
Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity
Sources
- CoreWeave Forge: pricing (checked 2026-11-15)
- Fireworks AI: pricing (checked 2026-11-14)
- DeepInfra: pricing (checked 2026-11-14)
- Featherless: pricing (checked 2026-11-15)
- Microsoft Azure: pricing (checked 2026-11-14)
- DeepInfra: pricing (checked 2026-11-14)
- Featherless: pricing (checked 2026-11-15)
- Microsoft Azure: pricing (checked 2026-11-14)
- Google Vertex AI: pricing (checked 2026-11-14)